In [1]:
import pymysql
import pandas as pd
import numpy as np
import seaborn as sns
from pandas import DataFrame, Series
import matplotlib.pyplot as plt
import sys
from datetime import datetime
# plotly packages and settings
import plotly
from plotly import tools
import plotly.graph_objs as go
import cufflinks as cf
#import modules to work with data visualizations offline
from plotly.offline import download_plotlyjs,init_notebook_mode,plot,iplot
#connect the Javascript to the notebook with the init_notebook_mode() method
# with the connected argument set to True
init_notebook_mode(connected=True)
#Update to use cufflinks offline
cf.go_offline()
pd.set_option('display.unicode.ambiguous_as_wide', True)
pd.set_option('display.unicode.east_asian_width', True)
from pyecharts import Line, Timeline, Grid,Bar,Scatter,EffectScatter
import math
from pyecharts import online
online()
import colorlover as cl
from IPython.display import HTML
import warnings
warnings.filterwarnings('ignore')
from pylab import mpl
mpl.rcParams['font.sans-serif'] = ['Fangsong'] # 指定默认字体
mpl.rcParams['axes.unicode_minus'] = False # 解决保存图像是负号'-'显示为方块的问题
In [2]:
def get_all_data(db,time_info):
###提取输入的数据库信息
host=db.split(':')[0]
port=int(db.split(':')[1].split('-')[0])
db_name=db.split('-')[-1]
#print(host,port,db_name)
#打开数据库连接
db = pymysql.connect(host = host, port = port, user = 'admin', \
passwd = 'Pwd@union9527', db = db_name, \
use_unicode=True, charset="utf8")
#创建游标对象
cursor = db.cursor()
sql = 'select * from lifetest_index'
cursor.execute(sql)
data = cursor.fetchall()
data = pd.DataFrame(list(data))
sql1 = 'select * from lifetest_daily_index'
cursor.execute(sql1)
data_day = cursor.fetchall()
data_day = pd.DataFrame(list(data_day))
db.close()
#添加列名
data_day.columns = ['day_laser','mean_day','max_day','min_day','range_day']
data_day['date'] = pd.to_datetime(data_day.day_laser.str.split('_').apply(lambda x:x[0]),format='%Y%m%d')
data_day['laser_id'] = data_day.day_laser.str.split('_').apply(lambda x:x[1])
data_day['server_id']=data_day.laser_id.apply(lambda x: '10.1.1.'+x[1:3])
data.columns=['filename','time','duration','avg_voltage','range_voltage',
'avg_current','range_current', \
'pwm','prop_0.1','prop_0.2',\
'prop_0.3','prop_0.3more','avg_duration','false_crc_prop',\
'uncomplete_frame_prop','missing_frame_prop','mean_hour',\
'max_hour','min_hour',\
'rand_err_trimed','max_daily_variation','std','range',\
'measure_5','measure_10','measure_15','measure_15more',\
'posi_rand_err','posi_rand_err_trimed','mean_hour_real_diff',\
'daily_mean_variation','error_interval','label']
data['laser_id'] = data.filename.str.split('-').apply(lambda x: x[-1]).str.split('.').apply(lambda x: x[0])
data['date'] = data.filename.str.split('-').apply(lambda x: x[0]).str.split('/').apply(lambda x: x[-1]).str.split('_').apply(lambda x: x[0])
data['date']=pd.to_datetime(data['date'],format='%Y%m%d.%H')
data['date2']=data.date.apply(lambda x:x.strftime("%Y%m%d.%H"))
data['date1'] = data['date'].apply(lambda x: x.date())
data['server_id']=data.filename.str.split('_').apply(lambda x:x[1]).str.split('-').apply(lambda x:x[0])
if time_info=='ALL':
return data,data_day
else:
time_start=datetime.strptime(time_info.split(',')[0]+' 00:00:00','%Y-%m-%d %H:%M:%S')
time_end=datetime.strptime(time_info.split(',')[1]+' 23:00:00','%Y-%m-%d %H:%M:%S')
min_time=data.date.min()
if time_start < min_time:
print('现有数据最早时间为：{}!'.format(min_time))
data1 = data[(data['date']>=time_start)&(data['date']<=time_end)]
data_day1 = data_day[(data_day['date']>=time_start)&(data_day['date']<=time_end)]
#data1=data1.fillna(0)#！！！有些项为空
return data1,data_day1
In [3]:
'''
def time_slice(data,data_day,time_info):
time_start=datetime.strptime(time_info.split(',')[0]+' 00:00:00','%Y-%m-%d %H:%M:%S')
time_end=datetime.strptime(time_info.split(',')[1]+' 23:00:00','%Y-%m-%d %H:%M:%S')
min_time=data.date.min()
if time_start < min_time:
print('现有数据最早时间为：{}!'.format(min_time))
data1 = data[(data['date']>=time_start)&(data['date']<=time_end)]
data_day1 = data_day[(data_day['date']>=time_start)&(data_day['date']<=time_end)]
return data1,data_day1
''';
In [4]:
#字典内容分别为：中文名称，纵坐标单位，(阈值下界，阈值上界)
dic = {'avg_current':['平均电压','单位(V)',11,13],\
'range_current':['电压稳定度','单位(V)',-200,200], \
'avg_voltage':['平均电流','单位(mA)'],\
'range_voltage':['电流稳定度','单位(mA)'],\
'pwm':['PWM值','单位(mA)',-100,100],\
'prop_0.1':['区间[66.5,66.7]占比','单位(%)'],\
'prop_0.2':['区间[66.4,66.5)和(66.7,66.8]占比','单位(%)'],\
'prop_0.3':['区间[66.3,66.5)和(66.8,66.9]占比','单位(%)'],\
'prop_0.3more':['小于66.3ms或大于66.9ms占比','单位(%)'],\
'max_hour':['小时最大值','单位(mm)'],\
'min_hour':['小时最小值','单位(mm)'],\
'avg_duration':['平均每帧用时','单位(ms)',65.66,67.66],\
'false_crc_prop':['CRC校验错误率','单位(%)',0.01],\
'uncomplete_frame_prop':['不完整帧占比','单位(%)',0.5],\
'missing_frame_prop':['漏帧率','单位(%)',0.5],\
'mean_hour':['小时平均值','单位(mm)'],\
'rand_err_trimed':['测量值随机误差最大值','单位(mm)',-30,30],\
'max_daily_variation':['日波动最大值','单位(mm)',-30,30],\
'std':['测量值标准差','单位(mm)'],\
'range':['测量值极差','单位(mm)',60],\
'measure_5':['$\\text{区间}[\\mu-5,\\mu+5]\\text{内占比}$','单位(%)'],\
'measure_10':['$\\text{区间}[\mu-10,\mu-5)\cup(\mu+5,\mu+10]\\text{内占比}$','单位(%)'],\
'measure_15':['$\\text{区间}[\mu-15,\mu-10)\cup(\mu+10,\mu+15]\\text{内占比}$','单位(%)'],\
'measure_15more':['$\\text{区间}(-\infty,\mu-15)\cup(\mu+15,+\infty]\\text{内占比}$','单位(%)'],\
'posi_rand_err':['随机误差最大值(未去异常值)出现的角度'],\
'posi_rand_err_trimed':['随机误差最大值(去异常值)出现的角度'],\
'mean_hour_real_diff':['小时平均值与真实值之差','单位(mm)',-10,10],\
'daily_mean_variation':['日均值波动','单位(mm)',10],\
'error_interval':['产品误差区间','单位(mm)',-20,20],
}
# 定义接口参数
para_dic={'VC':['avg_voltage','range_voltage',\
'avg_current','range_current'],\
'MO':['pwm','prop_0.1','prop_0.2',\
'prop_0.3','prop_0.3more','avg_duration'],\
'TR':['false_crc_prop',\
'uncomplete_frame_prop','missing_frame_prop'],\
'ST':[['mean_hour','max_hour','min_hour','mean_day','max_day','min_day'],\
['max_daily_variation','std','range','range_day',\
'measure_5','measure_10','measure_15','measure_15more','posi_rand_err_trimed'],\
['rand_err_trimed','mean_hour_real_diff','daily_mean_variation','error_interval']]}
In [5]:
##### 单个设备动态折线图 无阈值 #####
def argument_plot_no_threshold(df,argu):#argu:要可视化的判据(列名)
la = label_dic[argu]
df_yes = df[df[la]=="0"]
groups = df_yes.groupby('laser_id')
d1 = groups[argu].max()
d2 = groups[argu].min()
d3 = groups[argu].mean(numeric_only=False)
d_all = DataFrame([d1,d2,d3]).T
d_all.columns = ['最大值','最小值','平均值']
#d_all.columns = ['Max','Min','Average']
d_all.sort_values(by='最大值',ascending=False,inplace=True)
print(d_all)
df1 = df_yes.loc[:,[argu,'laser_id','date']].reset_index(drop=True).sort_values(by='date')
df2 = df1.pivot(index='date',columns='laser_id',values=argu)
_data = df2.iplot(kind='scatter',mode='lines+markers',size=4,opacity=0.5,
title=dic[argu][0],yTitle=dic[argu][1],asFigure=True)
_data['layout']['width'] = 600
_data['layout']['height'] = 300
_data['layout']['margin'] = dict(t=80,b=50,l=60)
_data.iplot()
#return d_all
##### 单个设备动态折线图 带阈值 #####
def argument_plot_with_threshold(df,argu):#输入需要可视化的判据
la = label_dic[argu]
df_yes = df[df[la]=="0"]
groups = df_yes.groupby('laser_id')
d1 = groups[argu].max()
d2 = groups[argu].min()
d3 = groups[argu].mean(numeric_only=False)
d_all = DataFrame([d1,d2,d3]).T
d_all.columns = ['最大值','最小值','平均值']
#d_all.columns = ['Max','Min','Average']
d_all.sort_values(by='最大值',ascending=False,inplace=True)
print(d_all)
df1 = df_yes.loc[:,[argu,'laser_id','date']].reset_index(drop=True).sort_values(by='date')
df2 = df1.pivot(index='date',columns='laser_id',values=argu)
_data = df2.iplot(kind='scatter',mode='lines+markers',size=4,opacity=0.5,
title=dic[argu][0],yTitle=dic[argu][1],asFigure=True)
_data['layout']['width'] = 600
_data['layout']['height'] = 300
_data['layout']['margin'] = dict(t=80,b=50,l=60)
_data['layout']['shapes'] = [{'type':'line',
'x0':df2.index[0],
'y0':dic[argu][2],
'x1':df2.index[-1],
'y1':dic[argu][2],
'line':{'color': 'rgb(50, 171, 96)',
'width': 2,
'dash':'dashdot' },
},
{'type':'line',
'x0':df2.index[0],
'y0':dic[argu][3],
'x1':df2.index[-1],
'y1':dic[argu][3],
'line':{'color': 'rgb(50, 171, 96)',
'width': 2,
'dash':'dashdot' },
}]
_data.iplot()
return d_all
##### 按照串口分组的折线图 #####
def server_line_plot(df,col,server):
la = label_dic[col]
data = df[df[la]=="0"]
num_server=len(server)
subplot_title = ['串口'+ser for ser in server]
row=int(np.ceil(num_server/2))
fig = tools.make_subplots(rows=row, cols=2, print_grid=False,\
shared_xaxes=True,subplot_titles=subplot_title)
i=2
j=1
for ser in server:
p=i//2;q=(j+1)%2+1
df = data[data['server_id']==ser]
df1 = df.loc[:,[col,'laser_id','date']].reset_index(drop=True).sort_values(by='date')
df2 = df1.pivot(index='date',columns='laser_id',values=col)
for co in df2.columns:
fig.append_trace({'x':df2.index,
'y':df2[co],
'type':'scatter',
'mode':'lines+markers',
'name':co},p,q)
i+=1
j+=1
fig['layout']['width']=800
fig['layout']['height']=300*row
fig['layout']['title']=dic[col][0]
fig['layout']['yaxis']=dict(title=dic[col][1])
iplot(fig)
##### 占比面积图 按设备分 #####
def prop_with_subplot(data,col,col_list,name):
subplot_title = ['设备'+la for la in laser_list]
fig = tools.make_subplots(rows=int(np.ceil(num_laser/2)), cols=2, \
vertical_spacing = 0.1,print_grid=False,
shared_xaxes=True,\
subplot_titles=subplot_title)
i=2
j=1
color =['rgb(184, 247, 212)','rgb(111, 231, 219)','rgb(127, 166, 238)','rgb(131, 90, 241)']
legend_group = ['group0','group1','group2','group3']
yaxis = ['yaxis'+str(i) for i in range(1,len(laser_list)+1)]
s=0
for la in laser_list:
p=i//2;q=(j+1)%2+1
df = data[data['laser_id']==la]
data_test = df.loc[:,['laser_id','prop_0.1','prop_0.2','prop_0.3','prop_0.3more','date']]
#data_test['date'] = data_test.date.apply(lambda x: x.strftime('%m%d.%H'))
#data_test['date'] = pd.to_datetime(data_test['date'],format='%m%d%Y.%H')
if (p==1)&(q==1):
for k in range(4):
trace=go.Scatter(
x=data_test['date'],
y=data_test[col_list[k]],
name=name[k],
legendgroup=legend_group[k],
showlegend=True,
mode='lines',
line=dict(width=0.5,
color=color[k]),
stackgroup='one',
)
fig.append_trace(trace,p,q)
else:
for k in range(4):
trace=go.Scatter(
x=data_test['date'],
y=data_test[col_list[k]],
name=name[k],
legendgroup=legend_group[k],
showlegend=False,
mode='lines',
line=dict(width=0.5,
color=color[k]),
stackgroup='one',
)
fig.append_trace(trace,p,q)
fig['layout'][yaxis[s]].update(ticksuffix='%')
i+=1
j+=1
s+=1
fig['layout']['width']=800
fig['layout']['height']=600
fig['layout']['title']=col
iplot(fig)
###### 测量值分布图 ######
def get_measure_data(data,col):
df = data[col].str.split(' ',expand=True)
df.columns=angle1
df['laser_id']=data['laser_id']
df['date']=data['date'].apply(lambda x:x.strftime(format='%m%d.%H'))
df['server_id']=data['server_id']
return df
def get_angle_df(measure5,measure10,measure15,measure15more,angle):
s1 = measure5[angle]
s1.name='measure_5'
s2 = measure10[angle]
s2.name='measure_10'
s3 = measure15[angle]
s3.name='measure_15'
s4 = measure15more[angle]
s4.name='measure_15more'
s5 = measure5['laser_id']
s6 = measure5['date']
s7 = measure5['server_id']
df = pd.concat([s1,s2,s3,s4,s5,s6,s7],axis=1)
return df
def prop_with_subplot_measure(data,la,col_list,name):
da = data[data['laser_id']==la]
subplot_title = ['角度'+str(an) for an in angle_list]
fig = tools.make_subplots(rows=1, cols=4, \
horizontal_spacing = 0.1,print_grid=False,
shared_yaxes=True,\
subplot_titles=subplot_title)
i=1
j=1
color =['rgb(184, 247, 212)','rgb(111, 231, 219)','rgb(127, 166, 238)','rgb(131, 90, 241)']
legend_group = ['group0','group1','group2','group3']
yaxis = ['yaxis'+str(i) for i in range(1,len(angle_list)+1)]
s=0
for an in angle_list:
data_test = get_angle_df(an)
if (i==1)&(j==1):
for k in range(4):
trace=go.Scatter(
x=data_test['date'],
y=data_test[col_list[k]],
name=name[k],
legendgroup=legend_group[k],
showlegend=True,
mode='lines',
line=dict(width=0.5,
color=color[k]),
stackgroup='one',
)
fig.append_trace(trace,i,j)
else:
for k in range(4):
trace=go.Scatter(
x=data_test['date'],
y=data_test[col_list[k]],
name=name[k],
legendgroup=legend_group[k],
showlegend=False,
mode='lines',
line=dict(width=0.5,
color=color[k]),
stackgroup='one',
)
fig.append_trace(trace,i,j)
fig['layout']['yaxis'].update(ticksuffix='%')
j+=1
s+=1
fig['layout']['width']=800
fig['layout']['height']=300
fig['layout']['title']=la
iplot(fig)
def time_line_server(data,col):
timeline = Timeline(is_auto_play=False,timeline_bottom=-5)
max_ids = data['laser_id'].drop_duplicates().groupby(data['server_id']).count().max()
server_list=data.server_id.unique()
judge=dic[col][0]
for ser in server_list:
df = data.query('server_id == @ser').reset_index(drop = True).sort_values(by='date')
#df = df.merge(time_unique,left_on='time_label',right_on='time',\
# how='outer').sort_values(by = 'time').reset_index(drop=True)
line = 'line'+ser
line=Line('串口'+ser+'的'+judge+'趋势图',dic[col][1],title_color='#FF6600',title_text_size=14)
laser_ids = sorted(data[data.server_id==ser].laser_id.unique())
for laser_id in laser_ids:
df1 = df[df.laser_id==laser_id]
line.add(laser_id,df1['date2'].tolist(),\
df1[col].tolist(), \
is_smooth=True,xaxis_rotate=30,yaxis_max='dataMax',\
yaxis_min ='dataMin',legend_orient='vertical',
legend_pos= 'right')
if len(laser_ids)!=max_ids:
for i in range(max_ids-len(laser_ids)):
line.add(laser_id,df[df.laser_id==laser_ids[-1]]['date2'].tolist(),\
df[df.laser_id==laser_ids[-1]][col].tolist(), is_smooth=True,\
xaxis_rotate=30,yaxis_max='dataMax',yaxis_min ='dataMin',\
legend_orient='vertical',
legend_pos= 'right')
grid = Grid()
grid.add(line, grid_bottom="20%")
timeline.add(grid, ser)
return timeline
judges=['帧用时分布','测量值分布']
def time_line_laser(data,judge,name,col_list):
timeline = Timeline(is_auto_play=False,timeline_bottom=-5)
laser_list1 = data.laser_id.unique()
for la in laser_list1:
df = data.query('laser_id == @la').reset_index(drop = True).sort_values(by='date')
line = 'line'+la
line=Line('设备'+la+'的'+judge+'比例面积图','单位：%',title_color='#FF6600',title_text_size=14)
for i in range(len(col_list)):
line.add(name[i],df['date2'].tolist(),\
df[col_list[i]].tolist(), \
is_fill=True,
is_stack=True,
line_opacity=0.2,
area_opacity=0.4,
is_smooth=True,xaxis_rotate=30,yaxis_max=100,\
yaxis_min =0,legend_orient='vertical',
legend_pos= 'right')
grid = Grid()
grid.add(line, grid_bottom="20%")
timeline.add(grid, la)
return timeline
def mean_max_min_line(data1_today_91_mean,data1_today_91_max,data1_today_91_min,judge):#judge=['小时'，'历史趋势图']
timeline = Timeline(timeline_bottom=-5,width=1000,height=900)
angles = np.arange(-45,225,45)
laser_list1=data1_today_91_mean.laser_id.unique()
for laser_id in laser_list1:
kk=0
df_mean = data1_today_91_mean.query('laser_id == @laser_id').reset_index(drop = True).sort_values(by='date')
df_max = data1_today_91_max.query('laser_id == @laser_id').reset_index(drop = True).sort_values(by='date')
df_min = data1_today_91_min.query('laser_id == @laser_id').reset_index(drop = True).sort_values(by='date')
grid1 = Grid(height=900,width=1000)
for angle in angles:
kk+=1
line = 'line'+laser_id+str(angle)
if kk==1:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge[0]+'最大、最小、平均值'+judge[1],"单位(mm)",title_color='#FF6600',title_text_size=14,title_pos='0%')
elif kk==2:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge[0]+'最大、最小、平均值'+judge[1],"单位(mm)",title_color='#FF6600',title_text_size=14,title_pos='47%')
elif kk==3:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge[0]+'最大、最小、平均值'+judge[1],"单位(mm)",title_color='#FF6600',title_text_size=14,title_top='34%',title_pos='0%')
elif kk==4:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge[0]+'最大、最小、平均值'+judge[1],"单位(mm)",title_color='#FF6600',title_text_size=14,title_pos='47%',title_top='34%')
elif kk==5:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge[0]+'最大、最小、平均值'+judge[1],"单位(mm)",title_color='#FF6600',title_text_size=14,title_pos='0%',title_top='65%')
elif kk==6:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge[0]+'最大、最小、平均值'+judge[1],"单位(mm)",title_color='#FF6600',title_text_size=14,title_pos='47%',title_top='65%')
line.add('平均值',df_mean['date'].tolist(),df_mean[angle].tolist(), is_smooth=True,xaxis_rotate=30,yaxis_max='dataMax',yaxis_min ='dataMin',legend_orient='vertical',
legend_pos= 'right')
line.add('最大值',df_mean['date'].tolist(),df_max[angle].tolist(), is_smooth=True,xaxis_rotate=30,yaxis_max='dataMax',yaxis_min ='dataMin',legend_orient='vertical',
legend_pos= 'right',line_type ='dashed')
line.add('最小值',df_mean['date'].tolist(),df_min[angle].tolist(), is_smooth=True,xaxis_rotate=30,yaxis_max='dataMax',yaxis_min ='dataMin',legend_orient='vertical',
legend_pos= 'right',line_type ='dotted')
if kk==1:
grid1.add(line,grid_left='5%',grid_bottom='70%',grid_right='55%')
elif kk==2:
grid1.add(line,grid_bottom='70%',grid_left='50%')
elif kk==3:
grid1.add(line,grid_left='5%',grid_top='39%',grid_right='55%',grid_bottom='38%')
elif kk==4:
grid1.add(line,grid_top='39%',grid_left='50%',grid_bottom='38%')
elif kk==5:
grid1.add(line,grid_left='5%',grid_top='70%',grid_right='55%',grid_bottom='7%')
elif kk==6:
grid1.add(line,grid_top='70%',grid_left='50%',grid_bottom='7%')
timeline.add(grid1, laser_id)
return timeline
###### 随机误差最大值出现的角度 ######
def bar_plot(data2a):
data2a['posi_rand_err1'] = data2a.posi_rand_err.apply(lambda x:angle1[int(x)])
group1 = data2a.groupby('laser_id')
n = len(group1)//2
fig,axes=plt.subplots(n,2, figsize=(12,3*n))
i=0
j=0
for la, gro in group1:
p=i//2;q=(j+2)%2
gro['posi_rand_err1'].value_counts().plot.bar(ax=axes[p,q])
axes[p,q].set_title('Laser ID:{}'.format(la),y=1.09)
i += 1
j += 1
plt.subplots_adjust(bottom=.05, top=.95, hspace=.5)
plt.suptitle('随机误差最大值出现的角度统计图')
plt.show()
def polar(data, col):
la = label_dic[col]
data = data[data[la]=="0"]
df = data[col].str.split(' ', expand=True)
df = df.apply(pd.to_numeric, errors='ignore')
df['laser_id'] = data['laser_id']
group = df.groupby('laser_id')
n = len(group)//2
fig,axes=plt.subplots(n,2, subplot_kw=dict(polar=True), figsize=(10,4*n))
i=0
j=0
for la, gro in group:
num = group.size()[la]
for k in range(num):
axes[i//2,(j+2)%2].plot(angle,gro.iloc[k,:91],'b.',alpha=0.2)
k += 1
axes[i//2,(j+2)%2].set_title('Laser ID:{}'.format(la),y=1.08)
axes[i//2,(j+2)%2].set_rlabel_position(270)
i += 1
j += 1
plt.subplots_adjust(bottom=.05, top=.95, hspace=.3)
plt.suptitle('判据:{}的极坐标图'.format(dic[col][0]),y=1,fontsize=14)
def line_plot(data,col):
la = label_dic[col]
data = data[data[la]=="0"]
df = data[col].str.split(' ', expand=True)
df = df.apply(pd.to_numeric, errors='ignore')
df['laser_id'] = data['laser_id']
df['date'] = data['date']
group = df.groupby('laser_id')
n = len(group)//2
fig,axes=plt.subplots(n,2, figsize=(12,3*n), sharex=True, sharey=True)
degree = [0,90,180]#degrees to plot
i=0
j=0
for la, gro in group:
p=i//2;q=(j+2)%2
#axes[i//2,(j+2)%2].plot(x=gro.date,y=gro.gro.iloc[:,np.where(angle1==degree[0])[0][0]],'b.-',label='0')
axes[p,q].plot(gro.date,gro.iloc[:,np.where(angle1==degree[0])[0][0]],'b.-',linewidth=1,label='0')
axes[p,q].plot(gro.date,gro.iloc[:,np.where(angle1==degree[1])[0][0]],'r+-',linewidth=1,label='90')
axes[p,q].plot(gro.date,gro.iloc[:,np.where(angle1==degree[2])[0][0]],'g<-',linewidth=1,label='180')
axes[p,q].set_title('Laser ID:{}'.format(la),y=1.08)
handles, labels = axes[p,q].get_legend_handles_labels()
axes[p,q].grid(True,linestyle='--')
axes[p,q].set_xticklabels(gro.date.apply(lambda x: x.strftime('%m%d.%H')),rotation=45)
i += 1
j += 1
fig.legend(handles, labels, loc='upper left')
plt.subplots_adjust(bottom=.05, top=.95, hspace=.5)
#plt.grid(True,linestyle='--')
plt.suptitle('判据:{}的位置{}折线图'.format(dic[col][0],degree),y=1,fontsize=14)
return df
def py_polar(col):
mean_hour=get_measure_data(col)
polar_mean = mean_hour.groupby('laser_id').first()
#polar_mean = mean_hour.groupby('laser_id').apply(lambda x: x.sample(1))
data_list=[]
color=cl.scales['8']['div']['Spectral']
index_list=polar_mean.index
for i in range(len(index_list)):
da =go.Scatterpolar(
r=polar_mean.iloc[i,:91],
theta=polar_mean.columns[:-2],
mode='markers',
name=index_list[i],
marker=dict(color=color[i],opacity=0.8))
data_list.append(da)
layout=go.Layout(title=dic[col][0],showlegend=True)
fig=go.Figure(data=data_list,layout=layout)
iplot(fig)
def py_polar_hour(col):
df = get_measure_data(data,col)
df1 = df.groupby('laser_id').last()
da = []
colors=cl.scales['8']['div']['Spectral']
k=0
for idx, row in df1.iterrows():
trace=go.Scatterpolar(
r=row,
theta=df1.columns[:-2],
name=idx,
mode='markers',
marker = dict(color=colors[k])
)
da.append(trace)
k+=1
layout=go.Layout(title=dic[col][0])
fig=go.Figure(data=da,layout=layout)
iplot(fig)
dic_day={'mean_day':'日平均值','max_day':'日最大值','min_day':'日最小值','range_day':'日极差'}
def py_polar_daily(col):
df = get_daily_measure_data(data_day,col)
df1 = df.groupby('laser_id').last()
da = []
colors=cl.scales['8']['div']['Spectral']
k=0
for idx, row in df1.iterrows():
trace=go.Scatterpolar(
r=row,
theta=df1.columns[:-2],
name=idx,
mode='markers',
marker = dict(color=colors[k])
)
da.append(trace)
k+=1
layout=go.Layout(title=dic_day[col])
fig=go.Figure(data=da,layout=layout)
iplot(fig)
def get_daily_measure_data(data,col):
df = data[col].str.split(' ',expand=True)
df.columns=angle1
df['laser_id']=data['laser_id']
df['date']=data['date'].apply(lambda x: x.strftime('%m-%d'))
return df
angle_list=np.arange(-45,225,45)
def time_line_laser_multi_angle(data,col):
timeline = Timeline(is_auto_play=False,timeline_bottom=-5)
judge=dic[col][0]
da = get_measure_data(data,col)
laser_list1=data.laser_id.unique()
for la in laser_list1:
df = da.query('laser_id == @la').reset_index(drop = True).sort_values(by='date')
line = 'line'+la
line=Line('设备'+la+'的'+judge+'多角度折线图','单位：{}'.format(dic[col][1]),title_color='#FF6600',title_text_size=14)
for i in range(len(angle_list)):
line.add(str(angle_list[i])+'°',df['date'].tolist(),\
df[angle_list[i]].tolist(), \
#is_symbol_show=True,
line_opacity=0.7,\
is_smooth=True,xaxis_rotate=30,yaxis_max='dataMax',\
yaxis_min ='dataMin',is_legend_show=True,legend_orient='vertical',
legend_pos= 'right')
grid = Grid()
grid.add(line, grid_bottom="20%")
timeline.add(grid, la)
return(timeline)
def time_area_measure_multi_angle(data,measure_5,measure_10,measure_15,measure_15more,judge):#judge=['小时'，'历史趋势图']
timeline = Timeline(is_auto_play=False,timeline_bottom=-5,width=1000,height=900)
angles = np.arange(-45,225,45)
laser_list1=data.laser_id.unique()
for laser_id in laser_list1:
kk=0
df_5 = measure_5.query('laser_id == @laser_id').reset_index(drop = True).sort_values(by='date')
df_10 = measure_10.query('laser_id == @laser_id').reset_index(drop = True).sort_values(by='date')
df_15 = measure_15.query('laser_id == @laser_id').reset_index(drop = True).sort_values(by='date')
df_15more = measure_15more.query('laser_id == @laser_id').reset_index(drop = True).sort_values(by='date')
grid1 = Grid(height=900,width=1000)
for angle in angles:
df=get_angle_df(df_5,df_10,df_15,df_15more,angle)
kk+=1
line = 'line'+laser_id+str(angle)
if kk==1:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge,"单位(%)",title_color='#FF6600',title_text_size=14,title_pos='0%')
elif kk==2:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge,"单位(%)",title_color='#FF6600',title_text_size=14,title_pos='47%')
elif kk==3:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge,"单位(%)",title_color='#FF6600',title_text_size=14,title_top='34%',title_pos='0%')
elif kk==4:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge,"单位(%)",title_color='#FF6600',title_text_size=14,title_pos='47%',title_top='34%')
elif kk==5:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge,"单位(%)",title_color='#FF6600',title_text_size=14,title_pos='0%',title_top='65%')
elif kk==6:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge,"单位(%)",title_color='#FF6600',title_text_size=14,title_pos='47%',title_top='65%')
line.add('区间[u-5,u+5]',df['date'].tolist(),df['measure_5'].tolist(), is_smooth=True,is_fill=True,\
is_stack=True,line_opacity=0.2,\
area_opacity=0.4,xaxis_rotate=30,yaxis_max=100,yaxis_min =0,legend_orient='vertical',\
legend_pos= 'right')
line.add('区间[u-10,u-5)U(u+5,u+10]',df['date'].tolist(),df['measure_10'].tolist(), is_smooth=True,\
is_fill=True,is_stack=True,line_opacity=0.2,\
area_opacity=0.4,xaxis_rotate=30,yaxis_max=100,yaxis_min =0,legend_orient='vertical',\
legend_pos= 'right')
line.add('区间[u-15,u-10)U(u+10,u+15]',df['date'].tolist(),df['measure_15'].tolist(), is_smooth=True,is_fill=True,\
is_stack=True,line_opacity=0.2,\
area_opacity=0.4,xaxis_rotate=30,yaxis_max=100,yaxis_min =0,legend_orient='vertical',legend_pos= 'right')
line.add('小于u-15或大于u+15',df['date'].tolist(),df['measure_15more'].tolist(), is_smooth=True,is_fill=True,\
is_stack=True,line_opacity=0.2,\
area_opacity=0.4,xaxis_rotate=30,yaxis_max=100,yaxis_min =0,legend_orient='vertical',legend_pos= 'right')
if kk==1:
grid1.add(line,grid_left='5%',grid_bottom='70%',grid_right='55%')
elif kk==2:
grid1.add(line,grid_bottom='70%',grid_left='50%')
elif kk==3:
grid1.add(line,grid_left='5%',grid_top='39%',grid_right='55%',grid_bottom='38%')
elif kk==4:
grid1.add(line,grid_top='39%',grid_left='50%',grid_bottom='38%')
elif kk==5:
grid1.add(line,grid_left='5%',grid_top='70%',grid_right='55%',grid_bottom='7%')
elif kk==6:
grid1.add(line,grid_top='70%',grid_left='50%',grid_bottom='7%')
timeline.add(grid1, laser_id)
return timeline
def time_area_error_multi_angle(data,judge):#judge=['小时'，'历史趋势图']
df_error=data.error_interval.str.split('],',expand=True)
df_error=df_error.applymap(lambda x: x.split('[')[-1])
#分出上下界
df_error_upper=df_error.applymap(lambda x: x.split(',')[1])
df_error_lower=df_error.applymap(lambda x: x.split(',')[0])
#定义列
df_error_upper.columns=angle1
df_error_lower.columns=angle1
df_error_upper['laser_id']=data['laser_id']
df_error_upper['date']=data['date']
df_error_upper.sort_values(by='date',inplace=True)
df_error_upper['date']=data['date'].apply(lambda x: x.strftime(format='%m%d.%H'))
df_error_lower['laser_id']=data['laser_id']
df_error_lower['date']=data['date']
df_error_lower.sort_values(by='date',inplace=True)
df_error_lower['date']=data['date'].apply(lambda x: x.strftime(format='%m%d.%H'))
#合并函数
df1=df_error_upper.copy()
df2=df_error_lower.copy()
timeline = Timeline(is_auto_play=False,timeline_bottom=-5,width=1000,height=900)
angles = np.arange(-45,225,45)
laser_list1=data.laser_id.unique()
for laser_id in laser_list1:
kk=0
df_u = df1.query('laser_id == @laser_id').reset_index(drop = True).sort_values(by='date')
df_l = df2.query('laser_id == @laser_id').reset_index(drop = True).sort_values(by='date')
grid1 = Grid(height=900,width=1000)
for angle in angles:
kk+=1
line = 'line'+laser_id+str(angle)
if kk==1:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge,"单位(mm)",title_color='#FF6600',title_text_size=14,title_pos='0%')
elif kk==2:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge,"单位(mm)",title_color='#FF6600',title_text_size=14,title_pos='47%')
elif kk==3:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge,"单位(mm)",title_color='#FF6600',title_text_size=14,title_top='34%',title_pos='0%')
elif kk==4:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge,"单位(mm)",title_color='#FF6600',title_text_size=14,title_pos='47%',title_top='34%')
elif kk==5:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge,"单位(mm)",title_color='#FF6600',title_text_size=14,title_pos='0%',title_top='65%')
elif kk==6:
line=Line('设备'+laser_id+'的'+str(angle)+'度的'+judge,"单位(mm)",title_color='#FF6600',title_text_size=14,title_pos='47%',title_top='65%')
line.add('误差上界',df_u['date'].tolist(),df_u[angle].tolist(), is_smooth=True,is_fill=False,\
is_stack=False,line_opacity=0.2,\
area_opacity=0.4,xaxis_rotate=30,yaxis_max='dataMax',yaxis_min ='dataMin',legend_orient='vertical',\
legend_pos= 'right')
line.add('误差下界',df_l['date'].tolist(),df_l[angle].tolist(), is_smooth=True,\
is_fill=True,is_stack=False,line_opacity=0.2,\
area_opacity=0.4,xaxis_rotate=30,yaxis_max='dataMax',yaxis_min ='dataMin',legend_orient='vertical',\
legend_pos= 'right')
if kk==1:
grid1.add(line,grid_left='5%',grid_bottom='70%',grid_right='55%')
elif kk==2:
grid1.add(line,grid_bottom='70%',grid_left='50%')
elif kk==3:
grid1.add(line,grid_left='5%',grid_top='39%',grid_right='55%',grid_bottom='38%')
elif kk==4:
grid1.add(line,grid_top='39%',grid_left='50%',grid_bottom='38%')
elif kk==5:
grid1.add(line,grid_left='5%',grid_top='70%',grid_right='55%',grid_bottom='7%')
elif kk==6:
grid1.add(line,grid_top='70%',grid_left='50%',grid_bottom='7%')
timeline.add(grid1, laser_id)
return timeline
def py_polar_hour(data,col):
df = get_measure_data(data,col)
df1 = df.groupby('laser_id').last()
da = []
colors=cl.scales['8']['div']['Spectral']
k=0
for idx, row in df1.iterrows():
trace=go.Scatterpolar(
r=row,
theta=df1.columns[:-2],
name=idx,
mode='markers',
marker = dict(color=colors[k])
)
da.append(trace)
k+=1
layout=go.Layout(title=dic[col][0])
fig=go.Figure(data=da,layout=layout)
iplot(fig)
def polar1(data, col):
la = label_dic[col]
data = data[data[la]=="0"]
df = data[col].str.split(' ', expand=True)
df = df.apply(pd.to_numeric, errors='ignore')
df['laser_id'] = data['laser_id']
if df.laser_id.nunique()==1:
laser_name = df.laser_id.unique()[0]
fig,ax=plt.subplots(figsize=(10,6))
fig.patch.set_visible(False)
for idx,row in df.iterrows():
plt.polar(angle,row[:91],'b.',alpha=0.2)
plt.title('Laser ID:{}的{}极坐标图'.format(laser_name,col),color='red')
elif df.laser_id.nunique()==2:
group = df.groupby('laser_id')
fig,axes=plt.subplots(1,2, subplot_kw=dict(polar=True), figsize=(10,4))
i=0
for la, gro in group:
num = group.size()[la]
for k in range(num):
axes[i].plot(angle,gro.iloc[k,:91],'b.',alpha=0.2)
k += 1
axes[i].set_title('Laser ID:{}'.format(la),y=1.08)
axes[i].set_rlabel_position(270)
i += 1
plt.subplots_adjust(bottom=.05, hspace=.3)
plt.suptitle('判据:{}的极坐标图'.format(dic[col][0]),color='red',y=1.01,fontsize=18)
elif (df.laser_id.nunique()>2)&(df.laser_id.nunique()%2==0):
group = df.groupby('laser_id')
n = len(group)//2
fig,axes=plt.subplots(n,2, subplot_kw=dict(polar=True), figsize=(10,4*n))
i=0
j=0
for la, gro in group:
num = group.size()[la]
#print(i//2,(j+2)%2)
for k in range(num):
axes[i//2,(j+2)%2].plot(angle,gro.iloc[k,:91],'b.',alpha=0.2)
k += 1
axes[i//2,(j+2)%2].set_title('Laser ID:{}'.format(la),y=1.08)
axes[i//2,(j+2)%2].set_rlabel_position(270)
i += 1
j += 1
plt.subplots_adjust(bottom=.05, hspace=.3)
plt.suptitle('判据:{}的极坐标图'.format(dic[col][0]),color='red',y=1.01,fontsize=18)
else:
group = df.groupby('laser_id')
n = len(group)//2+1
fig,axes=plt.subplots(n,2, subplot_kw=dict(polar=True), figsize=(10,4*n))
i=0
j=0
for la, gro in group:
num = group.size()[la]
#print(i//2,(j+2)%2)
for k in range(num):
axes[i//2,(j+2)%2].plot(angle,gro.iloc[k,:91],'b.',alpha=0.2)
k += 1
axes[i//2,(j+2)%2].set_title('Laser ID:{}'.format(la),y=1.08)
axes[i//2,(j+2)%2].set_rlabel_position(270)
i += 1
j += 1
plt.subplots_adjust(bottom=.05, hspace=.3)
plt.suptitle('判据:{}的极坐标图'.format(dic[col][0]),color='red',y=1.01,fontsize=18)
def bar_plot1(data2a):
data2a['posi_rand_err1'] = data2a.posi_rand_err.apply(lambda x:angle1[int(x)])
if data2a.laser_id.nunique()==1:
laser_name=data2a.laser_id.unique()[0]
data2a['posi_rand_err1'].value_counts().plot.bar()
plt.title('设备{}随机误差最大值出现的角度统计图'.format(laser_name))
elif data2a.laser_id.nunique()==2:
group1 = data2a.groupby('laser_id')
fig,axes=plt.subplots(1,2, figsize=(12,3))
i=0
for la, gro in group1:
gro['posi_rand_err1'].value_counts().plot.bar(ax=axes[i])
axes[i].set_title('Laser ID:{}'.format(la),y=1.05)
i += 1
plt.subplots_adjust(bottom=.05, hspace=.5)
plt.suptitle('随机误差最大值出现的角度统计图',color='#00008B',fontsize=18,y=1.03)
elif (data2a.laser_id.nunique()>2)&(data2a.laser_id.nunique()%2==0):
group1 = data2a.groupby('laser_id')
n = len(group1)//2
fig,axes=plt.subplots(n,2, figsize=(12,3*n))
i=0
j=0
for la, gro in group1:
p=i//2;q=(j+2)%2
gro['posi_rand_err1'].value_counts().plot.bar(ax=axes[p,q])
axes[p,q].set_title('Laser ID:{}'.format(la),y=1.05)
i += 1
j += 1
plt.subplots_adjust(bottom=.05, hspace=.5)
plt.suptitle('随机误差最大值出现的角度统计图',color='#00008B',fontsize=18,y=1.03)
else:
group1 = data2a.groupby('laser_id')
n = len(group1)//2+1
fig,axes=plt.subplots(n,2, figsize=(12,3*n))
i=0
j=0
for la, gro in group1:
p=i//2;q=(j+2)%2
gro['posi_rand_err1'].value_counts().plot.bar(ax=axes[p,q])
axes[p,q].set_title('Laser ID:{}'.format(la),y=1.05)
i += 1
j += 1
plt.subplots_adjust(bottom=.05, hspace=.5)
plt.suptitle('随机误差最大值出现的角度统计图',color='#00008B',fontsize=18,y=1.03)
plt.show()
In [6]:
def result_VC(data,analyze_object):
print("\033[1;31m\t{}\033[0m\n".format('电压电流特性：'))
if analyze_object=='ALL':
for argu in para_dic['VC']:
argument_plot_no_threshold(data,argu)
elif 'L' in analyze_object:
laser_input=analyze_object.split(',')[1:]
df=data[data['laser_id'].isin(laser_input)]
for argu in para_dic['VC']:
argument_plot_no_threshold(df,argu)
else:
server_input=analyze_object.split(',')[1:]
df=data[data['server_id'].isin(server_input)]
for argu in para_dic['VC']:
server_line_plot(df,argu,server_input)
def result_MO(data,analyze_object):
print("\033[1;31m\t{}\033[0m".format('电机特性：'))
if analyze_object=='ALL':
argument_plot_no_threshold(data,'pwm')
argument_plot_no_threshold(data,'avg_duration')
time_line_MO=time_line_laser(data,'帧用时分布',name1,col_list1)
elif 'L' in analyze_object:
laser_input=analyze_object.split(',')[1:]
df=data[data['laser_id'].isin(laser_input)]
argument_plot_no_threshold(df,'pwm')
argument_plot_no_threshold(df,'avg_duration')
time_line_MO=time_line_laser(df,'帧用时分布',name1,col_list1)
else:
server_input=analyze_object.split(',')[1:]
df=data[data['server_id'].isin(server_input)]
server_line_plot(df,'pwm',server_input)
server_line_plot(df,'avg_duration',server_input)
time_line_MO=time_line_laser(df,'帧用时分布',name1,col_list1)
#time_line_MO.render()
#a,b,c,d,e,f,g,h=''
return time_line_MO#,a,b,c,d,e,f,g,h
def result_TR(data,analyze_object):
print("\033[1;31m\t{}\033[0m".format('传输接口特性：'))
if analyze_object=='ALL':
for argu in para_dic['TR']:
argument_plot_no_threshold(data,argu)
elif 'L' in analyze_object:
laser_input=analyze_object.split(',')[1:]
df=data[data['laser_id'].isin(laser_input)]
for argu in para_dic['TR']:
argument_plot_no_threshold(df,argu)
else:
server_input=analyze_object.split(',')[1:]
df=data[data['server_id'].isin(server_input)]
for argu in para_dic['TR']:
server_line_plot(df,argu,server_input)
def result_ST(data,analyze_object):
print("\033[1;31m\t{}\033[0m".format('统计特性：'))
if analyze_object=='ALL':
#print("\033[1;31m\t{}\033[0m".format('产品趋势性指标：'))
#小时均值最大最小值
mean_hour = get_measure_data(data,'mean_hour')
max_hour = get_measure_data(data,'max_hour')
min_hour = get_measure_data(data,'min_hour')
#mean_max_min_line(mean_hour,max_hour,min_hour,judge=['小时','小时历史趋势图'])
timeline_hour=mean_max_min_line(mean_hour,max_hour,min_hour,judge=['小时','小时历史趋势图'])
py_polar_hour(data,'mean_hour')
#日均值最大最小值
mean_day = get_daily_measure_data(data_day,'mean_day')
max_day = get_daily_measure_data(data_day,'max_day')
min_day = get_daily_measure_data(data_day,'min_day')
timeline_day = mean_max_min_line(mean_day,max_day,min_day,['日','历史趋势图'])
#print("\033[1;31m\t{}\033[0m".format('产品离散度指标：'))
#polar(data,'max_daily_variation')
#time_line_laser_multi_angle(data,'max_daily_variation')
timeline_max_daily_variation=time_line_laser_multi_angle(data,'max_daily_variation')
polar(data,'std')
#time_line_laser_multi_angle(data,'std')
timeline_std=time_line_laser_multi_angle(data,'std')
measure_5 = get_measure_data(data,'measure_5')
measure_10 = get_measure_data(data,'measure_10')
measure_15 = get_measure_data(data,'measure_15')
measure_15more = get_measure_data(data,'measure_15more')
#time_area_measure_multi_angle(data,measure_5,measure_10,measure_15,measure_15more,judge='测量值分布图')#judge=['小时'，'历史趋势图']
timeline_measure=time_area_measure_multi_angle(data,measure_5,measure_10,measure_15,measure_15more,judge='测量值分布图')#judge=['小时'，'历史趋势图']
bar_plot1(data)
#print("\033[1;31m\t{}\033[0m".format('产品误差指标：'))
#polar(data,'rand_err_trimed')
#time_line_laser_multi_angle(data,'rand_err_trimed')
timeline_rand = time_line_laser_multi_angle(data,'rand_err_trimed')
polar(data,'daily_mean_variation')
timeline_mean_diff=time_line_laser_multi_angle(data,'daily_mean_variation')
timeline_error_band=time_area_error_multi_angle(data,'误差上下界')
elif 'L' in analyze_object:
laser_input=analyze_object.split(',')[1:]
data_L=data[data['laser_id'].isin(laser_input)]
data_day_L=data_day[data_day['laser_id'].isin(laser_input)]
#print("\033[1;31m\t{}\033[0m".format('产品趋势性指标：'))
#小时均值最大最小值
mean_hour = get_measure_data(data_L,'mean_hour')
max_hour = get_measure_data(data_L,'max_hour')
min_hour = get_measure_data(data_L,'min_hour')
#mean_max_min_line(mean_hour,max_hour,min_hour,judge=['小时','小时历史趋势图'])
timeline_hour=mean_max_min_line(mean_hour,max_hour,min_hour,judge=['小时','小时历史趋势图'])
py_polar_hour(data_L,'mean_hour')
#日均值最大最小值
mean_day = get_daily_measure_data(data_day_L,'mean_day')
max_day = get_daily_measure_data(data_day_L,'max_day')
min_day = get_daily_measure_data(data_day_L,'min_day')
timeline_day = mean_max_min_line(mean_day,max_day,min_day,['日','历史趋势图'])
#print("\033[1;31m\t{}\033[0m".format('产品离散度指标：'))
#polar(data,'max_daily_variation')
#time_line_laser_multi_angle(data,'max_daily_variation')
timeline_max_daily_variation=time_line_laser_multi_angle(data_L,'max_daily_variation')
polar1(data_L,'std')
#time_line_laser_multi_angle(data,'std')
timeline_std=time_line_laser_multi_angle(data_L,'std')
measure_5 = get_measure_data(data_L,'measure_5')
measure_10 = get_measure_data(data_L,'measure_10')
measure_15 = get_measure_data(data_L,'measure_15')
measure_15more = get_measure_data(data_L,'measure_15more')
#time_area_measure_multi_angle(data,measure_5,measure_10,measure_15,measure_15more,judge='测量值分布图')#judge=['小时'，'历史趋势图']
timeline_measure=time_area_measure_multi_angle(data_L,measure_5,measure_10,measure_15,measure_15more,judge='测量值分布图')#judge=['小时'，'历史趋势图']
bar_plot1(data_L)
#print("\033[1;31m\t{}\033[0m".format('产品误差指标：'))
#polar(data,'rand_err_trimed')
#time_line_laser_multi_angle(data,'rand_err_trimed')
timeline_rand = time_line_laser_multi_angle(data_L,'rand_err_trimed')
polar1(data_L,'daily_mean_variation')
timeline_mean_diff=time_line_laser_multi_angle(data_L,'daily_mean_variation')
timeline_error_band=time_area_error_multi_angle(data_L,'误差上下界')
else:
server_input=analyze_object.split(',')[1:]
data_S=data[data['server_id'].isin(server_input)]
data_day_S=data_day[data_day['server_id'].isin(server_input)]
#print("\033[1;31m\t{}\033[0m".format('产品趋势性指标：'))
#小时均值最大最小值
mean_hour = get_measure_data(data_S,'mean_hour')
max_hour = get_measure_data(data_S,'max_hour')
min_hour = get_measure_data(data_S,'min_hour')
#mean_max_min_line(mean_hour,max_hour,min_hour,judge=['小时','小时历史趋势图'])
timeline_hour=mean_max_min_line(mean_hour,max_hour,min_hour,judge=['小时','小时历史趋势图'])
py_polar_hour(data_S,'mean_hour')
#日均值最大最小值
mean_day = get_daily_measure_data(data_day_S,'mean_day')
max_day = get_daily_measure_data(data_day_S,'max_day')
min_day = get_daily_measure_data(data_day_S,'min_day')
timeline_day = mean_max_min_line(mean_day,max_day,min_day,['日','历史趋势图'])
#print("\033[1;31m\t{}\033[0m".format('产品离散度指标：'))
#polar(data,'max_daily_variation')
#time_line_laser_multi_angle(data,'max_daily_variation')
timeline_max_daily_variation=time_line_laser_multi_angle(data_S,'max_daily_variation')
polar1(data_S,'std')
#time_line_laser_multi_angle(data,'std')
timeline_std=time_line_laser_multi_angle(data_S,'std')
measure_5 = get_measure_data(data_S,'measure_5')
measure_10 = get_measure_data(data_S,'measure_10')
measure_15 = get_measure_data(data_S,'measure_15')
measure_15more = get_measure_data(data_S,'measure_15more')
#time_area_measure_multi_angle(data,measure_5,measure_10,measure_15,measure_15more,judge='测量值分布图')#judge=['小时'，'历史趋势图']
timeline_measure=time_area_measure_multi_angle(data_S,measure_5,measure_10,measure_15,measure_15more,judge='测量值分布图')#judge=['小时'，'历史趋势图']
bar_plot1(data_S)
#print("\033[1;31m\t{}\033[0m".format('产品误差指标：'))
#polar(data,'rand_err_trimed')
#time_line_laser_multi_angle(data,'rand_err_trimed')
timeline_rand = time_line_laser_multi_angle(data_S,'rand_err_trimed')
polar1(data_S,'daily_mean_variation')
timeline_mean_diff=time_line_laser_multi_angle(data_S,'daily_mean_variation')
timeline_error_band=time_area_error_multi_angle(data_S,'误差上下界')
return timeline_hour,timeline_day,timeline_max_daily_variation,timeline_std,timeline_measure,timeline_rand,timeline_error_band#,timeline_mean_diff
In [7]:
'''
db='10.1.1.202:3306-TEST'
time_info='ALL'
para='MO,TR'
analyze_object='L,8411,8413'
'''
#获取txt参数
import os
path=os.getcwd()
with open(path+'\\'+'sys_para.txt','r') as f1:
sys_para=f1.read()
sys_para_list=sys_para.split(';')
db = sys_para_list[0]
time_info=sys_para_list[3]
para=sys_para_list[1]
analyze_object=sys_para_list[2]
#解析原始数据
data,data_day=get_all_data(db,time_info)
data = data[~data.label.isna()].reset_index(drop=True)
# 对应列名和突变标识
d = data.columns[np.r_[3:8,12:23,29:31]]
df_label = data.label.str.split(' ',expand=True)
df_label.columns=['L'+str(x) for x in list(range(18))]
data = pd.concat([data,df_label],axis=1)
la = df_label.columns.values
label_dic={}
for i in range(len(la)):
label_dic[d[i]]=la[i]
server_list = data.server_id.unique()
num_server = len(server_list)
laser_list = sorted(data.laser_id.unique())
num_laser = len(laser_list)
col_list1=['prop_0.1','prop_0.2','prop_0.3','prop_0.3more']
name1 = [dic['prop_0.1'][0],dic['prop_0.2'][0],dic['prop_0.3'][0],dic['prop_0.3more'][0]]
name2=[dic['measure_5'][0],dic['measure_10'][0],dic['measure_15'][0],dic['measure_15more'][0],]
col_list2=['measure_5','measure_10','measure_15','measure_15more']
n = 91*3
a = -45
b = a+n
angle1 = np.arange(a,b,3)
angle = [math.radians(a) for a in angle1]
def final_result1(data,para,analyze_object):
###提取判据指标信息，list
if ',' in para:
para_list=para.split(',')
else:
para_list=[para]
'''
if 'MO' not in para_list:
timeline_MO=''
elif 'ST' not in para_list:
timeline_hour,timeline_day,timeline_max_daily_variation,timeline_std,timeline_measure,timeline_rand,timeline_error_band=['']*7
elif ('MO' not in para_list)&('ST' not in para_list):
'''
timeline_MO,timeline_hour,timeline_day,timeline_max_daily_variation,timeline_std,timeline_measure,timeline_rand,timeline_error_band=['']*8
for pa in para_list:
if pa=='VC':
result_VC(data,analyze_object)
elif pa=='MO':
timeline_MO=result_MO(data,analyze_object)
elif pa=='TR':
result_TR(data,analyze_object)
else:
timeline_hour,timeline_day,timeline_max_daily_variation,timeline_std,timeline_measure,timeline_rand,timeline_error_band=result_ST(data,analyze_object)
return timeline_MO,timeline_hour,timeline_day,timeline_max_daily_variation,timeline_std,timeline_measure,timeline_rand,timeline_error_band
#db='10.1.1.202:3306-TEST'
#time_info='ALL'
#para='MO,TR'
#analyze_object='L,8411,8413'
timeline_MO,timeline_hour,timeline_day,timeline_max_daily_variation,timeline_std,timeline_measure,timeline_rand,timeline_error_band=final_result1(data,para,analyze_object)
In [8]:
def return_timeline(timeline1):
if timeline1!='':
return timeline1
return_timeline(timeline_MO)
Out[8]:
In [9]:
return_timeline(timeline_hour)
Out[9]:
In [10]:
return_timeline(timeline_day)
Out[10]:
In [11]:
return_timeline(timeline_max_daily_variation)
Out[11]:
In [12]:
return_timeline(timeline_std)
Out[12]:
In [13]:
return_timeline(timeline_measure)
Out[13]:
In [14]:
return_timeline(timeline_rand)
Out[14]:
In [15]:
return_timeline(timeline_error_band)
Out[15]:
In [16]:
'''
#data为截取了时间信息的数据，para为判据组名，analyze_object为分析对象
def final_result(data,para,analyze_object):
###提取判据指标信息，list
if ',' in para:
para_list=para.split(',')
else:
para_list=[para]
if 'VC' in para_list:
print("\033[1;31m\t{}\033[0m\n".format('电压电流特性：'))
if analyze_object=='ALL':
for argu in para_dic['VC']:
argument_plot_no_threshold(data,argu)
elif 'L' in analyze_object:
laser_input=analyze_object.split(',')[1:]
df=data[data['laser_id'].isin(laser_input)]
for argu in para_dic['VC']:
argument_plot_no_threshold(df,argu)
else:
server_input=analyze_object.split(',')[1:]
df=data[data['server_id'].isin(server_input)]
for argu in para_dic['VC']:
server_line_plot(df,argu,server_input)
#a,b,c,d,e,f,g,h=''
#return a,b,c,d,e,f,g,h
#电机特性
if 'MO' in para_list:
print("\033[1;31m\t{}\033[0m".format('电机特性：'))
if analyze_object=='ALL':
argument_plot_no_threshold(data,'pwm')
argument_plot_no_threshold(data,'avg_duration')
time_line_MO=time_line_laser(data,'帧用时分布',name1,col_list1)
elif 'L' in analyze_object:
laser_input=analyze_object.split(',')[1:]
df=data[data['laser_id'].isin(laser_input)]
argument_plot_no_threshold(df,'pwm')
argument_plot_no_threshold(df,'avg_duration')
time_line_MO=time_line_laser(df,'帧用时分布',name1,col_list1)
else:
server_input=analyze_object.split(',')[1:]
df=data[data['server_id'].isin(server_input)]
server_line_plot(df,'pwm',server_input)
server_line_plot(df,'avg_duration',server_input)
time_line_MO=time_line_laser(df,'帧用时分布',name1,col_list1)
#time_line_MO.render()
#a,b,c,d,e,f,g,h=''
return time_line_MO#,a,b,c,d,e,f,g,h
if 'TR' in para_list:
print("\033[1;31m\t{}\033[0m".format('传输接口特性：'))
if analyze_object=='ALL':
for argu in para_dic['TR']:
argument_plot_no_threshold(data,argu)
elif 'L' in analyze_object:
laser_input=analyze_object.split(',')[1:]
df=data[data['laser_id'].isin(laser_input)]
for argu in para_dic['TR']:
argument_plot_no_threshold(df,argu)
else:
server_input=analyze_object.split(',')[1:]
df=data[data['server_id'].isin(server_input)]
for argu in para_dic['TR']:
server_line_plot(df,argu,server_input)
#a,b,c,d,e,f,g,h=''
#return a,b,c,d,e,f,g,h
if 'ST' in para_list:
print("\033[1;31m\t{}\033[0m".format('统计特性：'))
if analyze_object=='ALL':
#print("\033[1;31m\t{}\033[0m".format('产品趋势性指标：'))
#小时均值最大最小值
mean_hour = get_measure_data(data,'mean_hour')
max_hour = get_measure_data(data,'max_hour')
min_hour = get_measure_data(data,'min_hour')
#mean_max_min_line(mean_hour,max_hour,min_hour,judge=['小时','小时历史趋势图'])
timeline_hour=mean_max_min_line(mean_hour,max_hour,min_hour,judge=['小时','小时历史趋势图'])
py_polar_hour(data,'mean_hour')
#日均值最大最小值
mean_day = get_daily_measure_data(data_day,'mean_day')
max_day = get_daily_measure_data(data_day,'max_day')
min_day = get_daily_measure_data(data_day,'min_day')
timeline_day = mean_max_min_line(mean_day,max_day,min_day,['日','历史趋势图'])
#print("\033[1;31m\t{}\033[0m".format('产品离散度指标：'))
#polar(data,'max_daily_variation')
#time_line_laser_multi_angle(data,'max_daily_variation')
timeline_max_daily_variation=time_line_laser_multi_angle(data,'max_daily_variation')
polar(data,'std')
#time_line_laser_multi_angle(data,'std')
timeline_std=time_line_laser_multi_angle(data,'std')
measure_5 = get_measure_data(data,'measure_5')
measure_10 = get_measure_data(data,'measure_10')
measure_15 = get_measure_data(data,'measure_15')
measure_15more = get_measure_data(data,'measure_15more')
#time_area_measure_multi_angle(data,measure_5,measure_10,measure_15,measure_15more,judge='测量值分布图')#judge=['小时'，'历史趋势图']
timeline_measure=time_area_measure_multi_angle(data,measure_5,measure_10,measure_15,measure_15more,judge='测量值分布图')#judge=['小时'，'历史趋势图']
bar_plot(data)
#print("\033[1;31m\t{}\033[0m".format('产品误差指标：'))
#polar(data,'rand_err_trimed')
#time_line_laser_multi_angle(data,'rand_err_trimed')
timeline_rand = time_line_laser_multi_angle(data,'rand_err_trimed')
polar(data,'daily_mean_variation')
timeline_mean_diff=time_line_laser_multi_angle(data,'daily_mean_variation')
timeline_error_band=time_area_error_multi_angle(data,'误差上下界')
elif 'L' in analyze_object:
laser_input=analyze_object.split(',')[1:]
data_L=data[data['laser_id'].isin(laser_input)]
data_day_L=data_day[data_day['laser_id'].isin(laser_input)]
#print("\033[1;31m\t{}\033[0m".format('产品趋势性指标：'))
#小时均值最大最小值
mean_hour = get_measure_data(data_L,'mean_hour')
max_hour = get_measure_data(data_L,'max_hour')
min_hour = get_measure_data(data_L,'min_hour')
#mean_max_min_line(mean_hour,max_hour,min_hour,judge=['小时','小时历史趋势图'])
timeline_hour=mean_max_min_line(mean_hour,max_hour,min_hour,judge=['小时','小时历史趋势图'])
py_polar_hour(data_L,'mean_hour')
#日均值最大最小值
mean_day = get_daily_measure_data(data_day_L,'mean_day')
max_day = get_daily_measure_data(data_day_L,'max_day')
min_day = get_daily_measure_data(data_day_L,'min_day')
timeline_day = mean_max_min_line(mean_day,max_day,min_day,['日','历史趋势图'])
#print("\033[1;31m\t{}\033[0m".format('产品离散度指标：'))
#polar(data,'max_daily_variation')
#time_line_laser_multi_angle(data,'max_daily_variation')
timeline_max_daily_variation=time_line_laser_multi_angle(data_L,'max_daily_variation')
polar1(data_L,'std')
#time_line_laser_multi_angle(data,'std')
timeline_std=time_line_laser_multi_angle(data_L,'std')
measure_5 = get_measure_data(data_L,'measure_5')
measure_10 = get_measure_data(data_L,'measure_10')
measure_15 = get_measure_data(data_L,'measure_15')
measure_15more = get_measure_data(data_L,'measure_15more')
#time_area_measure_multi_angle(data,measure_5,measure_10,measure_15,measure_15more,judge='测量值分布图')#judge=['小时'，'历史趋势图']
timeline_measure=time_area_measure_multi_angle(data_L,measure_5,measure_10,measure_15,measure_15more,judge='测量值分布图')#judge=['小时'，'历史趋势图']
bar_plot1(data_L)
#print("\033[1;31m\t{}\033[0m".format('产品误差指标：'))
#polar(data,'rand_err_trimed')
#time_line_laser_multi_angle(data,'rand_err_trimed')
timeline_rand = time_line_laser_multi_angle(data_L,'rand_err_trimed')
polar1(data_L,'daily_mean_variation')
timeline_mean_diff=time_line_laser_multi_angle(data_L,'daily_mean_variation')
timeline_error_band=time_area_error_multi_angle(data_L,'误差上下界')
else:
server_input=analyze_object.split(',')[1:]
data_S=data[data['server_id'].isin(server_input)]
data_day_S=data_day[data_day['laser_id'].isin(laser_input)]
#print("\033[1;31m\t{}\033[0m".format('产品趋势性指标：'))
#小时均值最大最小值
mean_hour = get_measure_data(data_S,'mean_hour')
max_hour = get_measure_data(data_S,'max_hour')
min_hour = get_measure_data(data_S,'min_hour')
#mean_max_min_line(mean_hour,max_hour,min_hour,judge=['小时','小时历史趋势图'])
timeline_hour=mean_max_min_line(mean_hour,max_hour,min_hour,judge=['小时','小时历史趋势图'])
py_polar_hour(data_S,'mean_hour')
#日均值最大最小值
mean_day = get_daily_measure_data(data_day_S,'mean_day')
max_day = get_daily_measure_data(data_day_S,'max_day')
min_day = get_daily_measure_data(data_day_S,'min_day')
timeline_day = mean_max_min_line(mean_day,max_day,min_day,['日','历史趋势图'])
#print("\033[1;31m\t{}\033[0m".format('产品离散度指标：'))
#polar(data,'max_daily_variation')
#time_line_laser_multi_angle(data,'max_daily_variation')
timeline_max_daily_variation=time_line_laser_multi_angle(data_S,'max_daily_variation')
polar1(data_S,'std')
#time_line_laser_multi_angle(data,'std')
timeline_std=time_line_laser_multi_angle(data_S,'std')
measure_5 = get_measure_data(data_S,'measure_5')
measure_10 = get_measure_data(data_S,'measure_10')
measure_15 = get_measure_data(data_S,'measure_15')
measure_15more = get_measure_data(data_S,'measure_15more')
#time_area_measure_multi_angle(data,measure_5,measure_10,measure_15,measure_15more,judge='测量值分布图')#judge=['小时'，'历史趋势图']
timeline_measure=time_area_measure_multi_angle(data_S,measure_5,measure_10,measure_15,measure_15more,judge='测量值分布图')#judge=['小时'，'历史趋势图']
bar_plot(data_S)
#print("\033[1;31m\t{}\033[0m".format('产品误差指标：'))
#polar(data,'rand_err_trimed')
#time_line_laser_multi_angle(data,'rand_err_trimed')
timeline_rand = time_line_laser_multi_angle(data_S,'rand_err_trimed')
polar1(data_S,'daily_mean_variation')
timeline_mean_diff=time_line_laser_multi_angle(data_S,'daily_mean_variation')
timeline_error_band=time_area_error_multi_angle(data_S,'误差上下界')
return timeline_hour,timeline_day,timeline_max_daily_variation,timeline_std,timeline_measure,timeline_rand,timeline_error_band#,timeline_mean_diff
''';
In [17]:
'''
from IPython.display import Javascript
#para=sys.argv[2]
#analyze_object=sys.argv[3]
para='VC,MO,TR'
analyze_object='ALL'
timeline_hour,timeline_day,timeline_max_daily_variation,timeline_std,timeline_measure,timeline_rand,timeline_error_band=['']*7
if 'ST' not in para:
final_result(data,para,analyze_object)
Javascript('IPython.notebook.execute_cells_below()')
else:
timeline_hour,timeline_day,timeline_max_daily_variation,timeline_std,timeline_measure,timeline_rand,timeline_error_band=final_result(data,para,analyze_object)
Javascript('IPython.notebook.execute_cells_below()')
''';
In [18]:
from IPython.display import HTML
HTML('''<script>
code_show=true;
function code_toggle() {
if (code_show){
$('div.input').hide();
} else {
$('div.input').show();
}
code_show = !code_show
}
$( document ).ready(code_toggle);
</script>
<form action="javascript:code_toggle()"><input type="submit" value="Click here to toggle on/off the raw code."></form>''')
Out[18]: